Photoionization-Efficiency Spectrum and Ionization ... - ACS Publications

Nov 20, 1997 - Laboratory for Extraterrestrial Physics (Code 690), NASA/Goddard Space Flight Center,. Greenbelt, Maryland 20071. Szu-Cherng Kuo,| ...
0 downloads 0 Views 95KB Size
846

J. Phys. Chem. A 1998, 102, 846-851

Photoionization-Efficiency Spectrum and Ionization Energy of the Cyanomethyl Radical CH2CN and Products of the N(4S) + C2H3 Reaction R. Peyton Thorn Jr.,† Paul S. Monks,‡ and Louis J. Stief*,§ Laboratory for Extraterrestrial Physics (Code 690), NASA/Goddard Space Flight Center, Greenbelt, Maryland 20071

Szu-Cherng Kuo,| Zhengyu Zhang,⊥ Stuart K. Ross,∇ and R. Bruce Klemm*,# BrookhaVen National Laboratory, Building 815, P.O. Box 5000, Upton, New York 11973-5000 ReceiVed: September 30, 1997; In Final Form: NoVember 20, 1997

Photoionization efficiency (PIE) spectra of the CH2CN radical were measured over the wavelength range λ ) 115-130 nm using a discharge-flow-photoionization mass spectrometer coupled to a dispersed synchrotronradiation source. The cyanomethyl radical was produced by the reaction F + CH3CN f CH2CN + HF, and the PIE spectrum displayed steplike behavior near threshold. From the half-rise point of the initial step, a value of 10.280 ( 0.010 eV was obtained for the adiabatic ionization energy (IE) of CH2CN based on five independent determinations. From a single measurement of the PIE spectrum and threshold for CD2CN, we obtain IE(CD2CN) ) 10.24 eV. The experimental result for CH2CN is compared with previous measurements, estimates, and calculations. The present PIMS study of the CH2CN radical provides experimental measurements of the adiabatic ionization energy that are simultaneously the most direct and the most precise available. For the reaction N(4S) + C2H3, the C2H2N radical product exhibits a PIE spectrum that may include CH2CN along with another species that has a gradual threshold that is at a considerably longer wavelength than the steplike threshold of CH2CN (derived from F + CH3CN). A possible source of this difference is the contribution from higher-energy C2H2N isomers and/or from excited CH2CN. In sharp contrast to the results for the N(4S) + C2H3 reaction, no signal attributable to an isomer of the C2D2N radical was observed from the N(4S) + C2D3 reaction. The C2H3N/C2D3N adducts from the N(4S) + C2H3/C2D3 reactions were also studied. The adduct was observed to be solely CH3CN for the N(4S) + C2H3 reaction, while for N(4S) + C2D3, the PIE spectrum appears to include significant contributions from both the lowest-energy isomer CD3CN and one or more higher-energy isomers.

Introduction The cyanomethyl radical, CH2CN, and the related molecule acetonitrile, CH3CN, are prominent constituents in a wide variety of complex systems including hydrocarbon combustion,1 thermal decomposition,2 tropospheric and stratospheric chemistry of the earth’s atmosphere,3,4 and the chemistry of both interstellar clouds and the atmospheres of the outer planets.5 It has recently been established that one or more isomers of the CH2CN radical and the CH3CN adduct molecule are primary products of the reaction of ground-state atomic nitrogen with the vinyl radical C2H3.6 The thermochemistry of CH2CN and CH3CN is complicated by the existence of several isomers for the C2H2N family of † NAS/NRC Resident Research Associate. Email: ysrpt@ lepvax.gsfc.nasa.gov. ‡ NAS/NRC Resident Research Associate. Present Address: Chemistry Department, University of Leicester, University Road, Leicester, LE1 7RH, England. Email: [email protected]. § Email: [email protected]. | Present address: TRW Antenna Products Center, Mail Station 201/ 2055, Redondo Beach, CA 90278. Email: [email protected]. ⊥ Research Associate. Present address: Philips Lighting, Nanjing, Peoples Republic of China. ∇ Visiting Research Associate. Present address: Protection and Decontamination Department, CBD Porton Down, Salisbury, Wiltshire SP4 0JQ, U.K. # Email: [email protected]. * To whom correspondence should be sent.

radicals and cations7 and the C2H3N family of molecules and cations.8-10 Of the six C2H2N radical isomers that could be expected as products of the N + C2H3 reaction, heats of formation of the neutral and ion are available for only three isomers.7 For the five C2H3N molecule isomers potentially formed as adducts in this same reaction, heats of formation are known for all the neutral species and for four of the molecule ions.8-10 Present estimates of the ionization energy (IE) of the CH2CN radical are based on three distinct methods as reported in three very recent publications: the difference between experimentally determined values for ∆fH°298(CH2CN+) and ∆fH°298(CH2CN);7 a direct measurement1 of IE(CH2CN) via moderate-resolution electron-impact mass spectrometry (EIMS); a calculation11 using more sophisticated methods and considerably larger basis sets than previously employed. Earlier estimates based on the first method are not reliable because of a problem with the value used for ∆fH°298(CH2CN+), which was based on the appearance energy of the radical ion formed by dissociative ionization of CH3CN. Holmes and Mayer7 have shown that, contrary to expectation, dissociative ionization of CH3CN produces the lowest-energy cyclic isomer of the ion and not the linear CH2CN+ ion. In this work we report the first direct determination of the CH2CN photoionization efficiency (PIE) spectrum and photo-

S1089-5639(97)03164-2 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/09/1998

Cyanomethyl Radical CH2CN

J. Phys. Chem. A, Vol. 102, No. 5, 1998 847

ionization threshold. From the latter we obtain the first direct, high-resolution measurement of the ionization energy of the CH2CN radical. Our result is compared with those from previous experimental and theoretical studies. We also report on the PIE spectra of the C2H2N radical and C2H3N adduct molecule formed in the reaction N(4S) + C2H3. Products of the N(4S) + C2D3 reaction are also investigated. Experimental Section Experiments were performed by employing a dischargeflow-photoionization mass spectrometer (DF-PIMS) apparatus coupled to beamline U11 at the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory. The apparatus and experimental procedures have been described in previous publications.12-14 The cyanomethyl radical was produced in a Teflon lined flow reactor by the reaction of atomic fluorine with an excess of acetonitrile:

F + CH3CN f CH2CN + HF

(1)

k1(298 K) ) 1.2 × 10-11 cm3 molecule-1 s-1 (ref 1) Fluorine atoms were produced by passing a dilute mixture of F2 (∼5% in He) through a microwave discharge ( 105 nm) was used to eliminate second- and higher-order radiation. For direct CH3CN ionization and for the C2H3N adduct product from N(4S) + C2H3, it was necessary to work below λ ) 105 nm, and thus, the LiF window was not in place. In this case, corrections were made for second-order light by scanning the spectral range at one-half the wavelength range and one-half the wavelength step size. These short-wavelength scans were renormalized to reflect the intensity of second-order radiation on the first-order scan and subtracted from the raw data.18 The intensity of the VUV light was monitored via a sodium salicylate coated window with an attached photomultiplier tube. Acetonitrile (99%, Aldrich Chemical Co.), perdeuterated acetonitrile (99.8% D, Cambridge Isotope Laboratory), ethylene (research grade, 99.5%, MG Industries), and perdeuterated ethylene (98% D, Cambridge Isotope Laboratory) were all thoroughly outgassed by repeated freeze-pump-thaw cycles at T ) 77 K. Helium (research grade, 99.9999%, MG Industries), nitrogen (99.999%, MG Industries), and fluorine (5% mixture of F2 in helium, Cryogenic Rare Gases and MG Industries) were all used as supplied. Results and Discussion The PIE spectrum of the source molecule CH3CN (m/z ) 41) is shown in Figure 1A as an example of the PIMS experiment as well as a check on the wavelength calibration as determined by the zero-order setting,12 which was adjusted at the beginning and checked at the end of each filling of the VUV ring. The ionization threshold, which is generally taken as the half-rise point of the first step (unless otherwise specified), is indicated in Figure 1B at λ ) 101.68 nm. This corresponds to IE(CH3CN) ) 12.194 ( 0.013 eV (where the uncertainty is conservatively estimated from the resolution), in excellent agreement with the recommended9 value 12.194 ( 0.005 eV, which is also based on a photoionization study.19 This level of agreement indicates that the wavelength calibration is reliable and that the threshold is not significantly perturbed by thermal effects. The structure evident in the PIE spectrum of CH3CN is a significant feature that is an unambiguous identifying characteristic or “fingerprint” of the molecule. This structure is related, according to Rider et al.,19 to an autoionizing Rydberg state that converges to an excited state of the cation at 13.13 eV (∼94.4 nm; see Figure 1A). A. Ionization Energy of the CH2CN Radical. The PIE spectrum for the CH2CN radical (m/z ) 40), produced in the reaction F + CH3CN, is shown in Figure 2. This spectrum was obtained in the wavelength region λ ) 115.0-125.0 nm at 0.20 nm intervals and a nominal resolution of 0.16 nm (fwhm). The somewhat restricted wavelength region is dictated by two

848 J. Phys. Chem. A, Vol. 102, No. 5, 1998

Figure 1. (A) Photoionization-efficiency spectrum of CH3CN between λ ) 90.0 and 105.0 nm at a nominal resolution of 0.16 nm with 0.1nm steps. The photoionization efficiency is the ion counts at m/z ) 41 divided by the light intensity in arbitrary units. [CH3CN] ) 5.0 × 1013 molecules cm-3. The structure in this spectrum, with apparent peaks at about 98.4, 97.1, and 96.4 nm, is associated with a Rydberg series (see text) that converges to a limit at 94.45 nm. (B) Photoionization threshold region of CH3CN between λ ) 100.0 and 105.0 nm at a nominal resolution of 0.11 nm with 0.05-nm steps. The splitting in the threshold is presumably due to Jahn-Teller coupling in the CCN bending vibration.19 The onset of ionization is at λ ) 101.68 nm (12.194 ( 0.013 eV).

Figure 2. Photoionization-efficiency spectrum and threshold of CH2CN between λ ) 115.0 and 125.0 nm. The onset of ionization is at λ ) 120.6 nm (10.281 ( 0.014 eV). [CH3CN]0 ) 2.1 × 1014 molecules cm-3 and [F2]0 ) 8.5 × 1012 molecules cm-3.

factors. The first is our preference to avoid, when possible, operating without the LiF window and therefore to avoid the need to correct for second-order light from the monochromator. This limits the shorter-wavelength limit to above the LiF cutoff, i.e., above the minimum λ ) 105 nm where the UV flux is greatly reduced even for a clean window not damaged via F-center formation. The second factor is the location of the photoionization threshold, which determines the longerwavelength limit to be scanned.

Thorn et al. There is only limited structure discernible in Figure 2, which may be due to either autoionizing Rydberg states and/or cation vibrational levels above n ) 0. From Figure 2 a threshold wavelength of λ ) 120.6 nm is obtained corresponding to IE(CH2CN) ) 10.281 eV. The results of five independent determinations of IE(CH2CN) that covered various wavelength ranges with wavelength step intervals of 0.1 or 0.2 nm and nominal resolutions of 0.13-0.16 nm yield a simple average of 10.280 ( 0.010 eV for IE(CH2CN) where the precision is at the 2σ level. From a single measure of the PIE spectrum and threshold for CD2CN (formed via the F + CD3CN reaction), we obtain a threshold wavelength of λ ) 121.1 nm. This wavelength corresponds to IE(CD2CN) ) 10.24 eV, which is 0.04 eV lower than the value for CH2CN. The CD2CN PIE spectrum is essentially the same as that for CH2CN shown in Figure 2. It must be emphasized that the rather sharp transition observed at the onset of ionization (Figure 2) for the CH2CN radical and the fact that the lowest state of the cation is the cyclic isomer7 mean that we are observing the transition to the linear CH2CN+ structure and not the transition to the lower-energy cyclic cation. The latter transition would have very low probability (vanishingly small Franck-Condon factors) due to the large change in geometry. A comparison of our result for IE(CH2CN) with previous estimates is presented in Table 1. These estimates include an earlier measurement20 via EIMS as well as a very recent one1 using the same technique. Upper and lower limits are available from PIMS studies that employed rare-gas resonance lamps;21,22 it was observed that CH2CN was readily detected and monitored with an Ar lamp/LiF window (11.6, 11.8 eV)21,22 but not with a Kr lamp/MgF2 window (10.0, 10.6 eV).22 A value for IE(CH2CN) may also be determined from the difference between the recently measured values for ∆fH(CH2CN+) and ∆fH(CH2CN).7,9 Finally, there is a very recent theoretical calculation that employed very large basis sets.11 It may be seen in Table 1 that our value of IE(CH2CN) ) 10.280 eV agrees best with the two previous estimates that should be the more accurate and reliable: 10.1 eV determined from the difference in the heats of formation of the cation and neutral7 and 10.20 eV calculated theoretically.11 The energy resolution in the two direct EIMS studies1,20 of the CH2CN free radical was at least an order of magnitude less than that in our PIMS study. The recent value1 (10.5 ( 0.3 eV) agrees with our more precise result within their stated uncertainty, while the older value20 (10.87 ( 0.10 eV) is too large by an amount that far exceeds the combined experimental uncertainties. This could be due to a significant error in the older measurements and/or a substantial underestimation of the experimental uncertainty. The resonancelamp PIMS studies do give satisfactory upper and lower limits to our result if it is recognized that the more energetic Kr resonance line at λ ) 116.5 nm (10.6 eV) is much weaker in a typical Kr resonance lamp (20% of the 10.0 eV line at λ ) 123.6 nm) and may be further attenuated by deposits on the MgF2 window. Thus, a better lower limit is probably IE(CH2CN) > 10.0 eV. A value for IE(CH2CN) cannot be obtained from the appearance energy of the C2H2N+ fragment ion from CH3CN because the fragment ion formed is the ground-state cyclic isomer.7 In summary, the present PIMS study of the CH2CN radical provides an experimental measurement of the adiabatic ionization energy of this species that is simultaneously the most direct and the most precise available. B. Free-Radical Products at m/z ) 40 and 42 of the N(4S) + C2H3/C2D3 Reactions. At a photon energy of 11.3 eV (λ )

Cyanomethyl Radical CH2CN

J. Phys. Chem. A, Vol. 102, No. 5, 1998 849

TABLE 1: Comparison of Values for IE(CH2CN) IE(CH2CN) (eV)

methoda

ref

10.87 ( 0.10 10.0-10.6 10.1 ( 0.2b 10.5 ( 0.3 10.20 ( 0.05 10.280 ( 0.010

EIMS PIMS/Ar lamp derived PIMS/Kr lamp ∆fH(CH2CN+) - ∆fH(CH2CN) EIMS ab initio calculation PIMS/synchrotron

Pottie and Lossing (1961)20 Park and Gutman (1983)21 Lias et al. (1993)9 Masaki et al. (1995)22 Holmes and Mayer (1995)7 Hoyermann and Seeba (1995)1 Horn et al. (1995)11 this study

a EIMS, electron impact mass spectrometry; PIMS, photoionization mass spectrometry. b This value was derived from heats of formation of the CH2CN radical and the cation at T ) 298 K and strictly should not be equated to the IE. However, the error in doing so is only the difference in the integrated heat capacities of CH2CN+ and CH2CN, which is probably about 1-2 kJ mol-1 (0.01-0.02 eV).

Figure 3. Photoionization-efficiency spectrum of C2H3F between λ ) 112.0 and 122.0 nm at a nominal resolution of 0.13 nm with 0.1-nm steps. The onset of ionization is at λ ) 119.7 nm (10.358 ( 0.011 eV). [C2H4]0 ) 7.7 × 1014 molecules cm-3 and [F2]0 ) 2.8 × 1013 molecules cm-3.

110 nm) in the N(4S) + C2H3 system, major peaks at m/z ) 40 and m/z ) 46 are observed. The feature at m/z ) 46 is present in the absence of N and is due to C2H3F formed along with C2H3 in the reaction of F with C2H416 (see reaction 2). The PIE spectrum of C2H3F formed in reaction 2b is shown in Figure 3. The photoionization threshold is at λ ) 119.7 nm. This corresponds to IE(C2H3F) ) 10.358 eV, which agrees well with the recommended9 value of 10.363 eV and again demonstrates that the wavelength calibration is reliable. We also note the absence of significant signal above background due to “hot bands”, which indicates that any excess internal energy (∆rH° -1 9 298 ) -52.7 kJ mol ) is rapidly lost via collisions in the flow tube. The PIE spectrum of the C2H2N radical (m/z ) 40) produced in the reaction N(4S) + C2H3 is shown in Figure 4 (curves “b” and “c”),

N(4S) + C2H3 f C2H2N + H

(3a)

This spectrum was obtained in the wavelength region λ ) 115.0-130.0 nm at 0.2-nm intervals with a nominal resolution of 0.17 nm. The spectrum (Figure 4, curves “b” and “c”) differs in two significant ways from that of the CH2CN radical isomer formed via F + CH3CN and recorded under similar experimental conditions (Figure 4, curve “a”). First, although the CH2CN spectrum shows a sharp, steplike transition at threshold, the C2H2N isomer (or isomers) formed via N(4S) + C2H3 displays a gradually decreasing signal as wavelength increases. Second, although the threshold for the radical product of N(4S) + C2H3

Figure 4. (a) Curve (O) is the photoionization-efficiency spectrum of CH2CN radical from Figure 2. (b) Curve (b) is the photoionizationefficiency spectrum of C2H2N radical (formed in the N(4S) + C2H3 reaction) between λ ) 115.0 and 130.0 nm. [C2H4]0 ) 5.3 × 1014 molecules cm-3, [F2]0 ) 2.7 × 1013 molecules cm-3, and [N2]0 ) 1.9 × 1015 molecules cm-3. (c) Curve (4) is the same as curve in (b) expanded by a factor of 10 to highlight the threshold region.

Figure 5. Enthalpies of formation and ionization energies of CH2CN, CH2NC, and cyclic C2H2N (see text for details and references).

is difficult to define, it is at a considerably longer wavelength (λ > 128.6 nm, IE < 9.64 eV) than the sharp threshold observed for the CH2CN isomer (λ ) 120.6 nm, IE ) 10.280 eV). The obvious conclusion is that the C2H2N radical product from the N(4S) + C2H3 reaction may not be solely identified with the equilibrated linear ground-state isomer CH2CN. The enthalpies of formation7,9 and IEs1,7,11 of CH2CN, CH2NC, and cyclic C2H2N are summarized in Figure 5. Although the IEs for the CH2NC isomer (IE ) 9.4 eV, λ ) 132 nm)7 and the cyclic isomer N HC

CH

850 J. Phys. Chem. A, Vol. 102, No. 5, 1998

Figure 6. Upper curve (b) is the photoionization-efficiency spectrum of CH3CN formed in the N(4S) + C2H3 reaction between λ ) 90.0 and 105.0 nm. [C2H4]0 ) 4.1 × 1014 molecules cm-3, [F2]0 ) 3.5 × 1013 molecules cm-3, and [N2]0 ) 1.1 × 1015 molecules cm-3. Lower curve (O) is the photoionization-efficiency spectrum of CH3CN from Figure 1A.

(IE < 8.3 eV, λ > 149 nm)7 do not appear to match that for the species detected in this work, there are no reported PIE spectra for these isomers, and therefore, their photoionization threshold behavior is unknown. The IEs of the three other C2H2N isomers, H HC C

Thorn et al.

Figure 7. Upper curve (b) is the photoionization-efficiency spectrum of C2D3N formed in the N(4S) + C2D3 reaction between λ ) 90.0 and 140.0 nm. [C2D4]0 ) 3.7 × 1014 molecules cm-3, [F2]0 ) 3.5 × 1013 molecules cm-3, and [N2]0 ) 1.1 × 1015 molecules cm-3. Lower curve (O) is the photoionization-efficiency spectrum of CD3CN between λ ) 90.0 and 110.0 nm at the same resolution and step size as those for the upper trace. [CD3CN]0 ) 1.1 × 1014 molecules cm-3.

the C2H3N adduct closely matches that of acetonitrile. The IEs of three of the remaining C2H3N isomers are known to be considerably lower than our observed value of 12.20 eV, i.e., 11.2 eV for CH3NC,9 about 10.1 eV for the cyclic isomer 2Hazarine8,9

N, HC C NH, H2C C N N

are unknown as are their heats of formation. Thus, the potential contribution of these isomers to the C2H2N product spectrum in curves b and c of Figure 4 cannot be evaluated at present. A contribution from thermally excited CH2CN, which would extend the threshold to lower energies as observed, cannot be ruled out. However, our experience has been that rapid thermalization of reaction products is generally achieved in the flow reactor. For example, in the present study there is no compelling evidence for thermal excitation in the PIE spectra of either the C2H3F product (Figure 3) or the CH3CN product (Figure 6), both of which are formed initially “hot”. We are therefore unable to identify the C2H2N product at this time. Further studies on both the C2H2N radical product of the N(4S) + C2H3 reaction and on the thermochemical properties of all the C2H2N radical isomers are clearly required. A search for the PIE signal of the expected C2D2N radical (m/z ) 42) from the reaction N(4S) + C2D3 proved to be entirely negative. Under conditions very similar to these employed above for detection of C2H2N from N(4S) + C2H3, no net signal above background was detected. This indicates a yield of less than 1% for C2D2N compared to a yield of 80% (ref 6) for C2H2N. We have no explanation at this time for such a strong and unexpected isotope effect. C. Adduct Molecule Products at m/z ) 41 and 44 of the N(4S) + C2H3/C2D3 Reactions. The PIE spectrum of the C2H3N adduct molecule (m/z ) 41) formed in the N(4S) + C2H3 reaction is shown in Figure 6 (upper curve). This spectrum was measured over the wavelength region λ ) 90.0-105.0 nm at 0.2-nm intervals with a nominal resolution of 0.10 nm. The ionization threshold is at λ ) 101.6 nm (12.20 eV). This result is in excellent agreement with the PIE threshold for CH3CN shown in the lower curve in Figure 6 (λ ) 101.68 nm, IE ) 12.194 eV) and the recommended9 value of IE(CH3CN) ) 12.194 eV. Additionally, the structure in the PIE spectrum of

HC

CH2

and ∼8.3 eV for H2CdCdNH (ketene imine).6,9,10 The IE is unknown for the isomer vinyl nitrene H N

C CH2

We therefore conclude that the lowest-energy isomer, CH3CN, is the major and probably exclusive C2H3N adduct isomer (m/z ) 41) formed in the N(4S) + C2H3 reaction under the experimental conditions of the present photoionization study and the previous electron-impact study.6 The identity of the adduct molecule is much less clear for the N(4S) + C2D3 reaction. The PIE spectrum of the C2D3N adduct molecule (m/z ) 44) is shown in the upper curve of Figure 7. The spectrum was obtained in the wavelength region 90.0-140.0 nm at 0.2-nm intervals and a nominal resolution of 0.10 nm. The difference between this spectrum and that obtained for deuterated acetonitrile (CD3CN) under the same conditions (Figure 7, lower curve) is immediately obvious and striking. In the region λ ) 90-100 nm, both spectra exhibit the same fingerprint structure and a generally decreasing signal with increasing wavelength. But at wavelengths larger than 100 nm, the authentic CD3CN signal continues to decrease with a well-defined step threshold at λ ) 101.4 nm. In contrast, the C2D3N spectra from the N(4S) + C2D3 reaction exhibits a signal increase above λ ) 100 nm, reaching a maximum near λ ) 106 nm and then decreasing rather continuously and gradually to a threshold at approximately λ ) 129.2 nm (9.596 eV). In the C2D3N spectrum (Figure 7, upper curve) the structure in the region λ ) 90-100 nm due to CD3CN and the unidentified structure in the region λ ) 100-125 nm was confirmed by replicate runs. The 9.596-eV threshold does not correspond to

Cyanomethyl Radical CH2CN

J. Phys. Chem. A, Vol. 102, No. 5, 1998 851

the IE of the nondeuterated versions of the isomers N CD3NC, DC

CD2, or D2C C

ND

but only the ketene imine isomer may be excluded from contribution. We therefore conclude that the C2D3N adduct formed in the reaction N(4S) + C2D3 includes a significant contribution from the lowest-energy isomer CD3CN (acetonitrile), as well as possible contributions from

U.S. Department of Energy (under Contract No. DE-AC02786CH00016) and the Laboratory Directed Research and Development Program at Brookhaven National Laboratory. The work at GSFC was supported by the NASA Planetary Atmospheres Research Program and the NASA Upper Atmosphere Research Program. R.P.T. and P.S.M. thank the NAS/NRC for the award of a research associateship. References and Notes (1) Hoyermann, K.; Seeba, J. Z. Phys. Chem. 1995, 188, 215. (2) Lifshitz, A.; Moran, A.; Bidasic, S. Int. J. Chem. Kinet. 1987, 19,

N CD3NC, DC

CD2

and perhaps D N

C CD2

(vinyl nitrene) or other unidentified isomers. A more definitive conclusion must await information on the IEs of such isomers. The difference in the adduct isomers observed for the N(4S) + C2H3 and N(4S) + C2D3 reactions may be related to the fate of the initially formed adduct, which is very likely 2H-azarine if N adds across the CdC double bond (to form the cyclic isomer) or else vinyl nitrene if N attacks the C atom having only one H attached (to form the biradical). Subsequent isomerization of either of these isomers to the lowest-energy CH3CN isomer involves an H-atom transfer to the adjacent carbon.10 For the present reaction conditions (∼7-ms reaction time), this process is facile for the protonated species but inhibited for the deuterated species. This behavior is consistent with a mechanism that involves a 1,2-H/D shift, the kinetics of which are controlled by tunneling. Notes Added in Proof. After the present paper was accepted for publication, a photoelectron spectroscopy (PES) study was published (Shea, D. A.; Steenvoorden, J. J. M.; Chen, P. J. Phys. Chem. A 1997, 101, 9728) that reports a value of 10.30 ( 0.04 eV for the adiabatic ionization energy of CH2CN radical. In that study, CH2CN was generated via the thermal dissociation of “chloro-, bromo-, or iodoacetonitrile ... at temperatures >1200 K (10-ms contact time)”. Since the PES study employed both a different measurement technique and a different radicalgeneration procedure, we believe that the good agreement between our two studies demonstrates the definitive determination of IE(CH2CN). Acknowledgment. The work at BNL was supported by the Chemical Sciences Division, Office of Basic Energy Sciences,

61. (3) Kurylo, M. J.; Knable, G. L. J. Phys. Chem. 1984, 88, 3305. (4) Poulet, G.; Laverdet, J. L.; Jourdain, J. L.; LeBras, G. J. Phys. Chem. 1984, 88, 6259. (5) (a) Irvine, W. M.; Friberg, P.; Hjalmarson, A.; Ishikawa, S.; Kaifu, N.; Kawaguchi, K.; Madden, S. C.; Matthews, H. E.; Ohishi, M.; Saito, S.; Suzuki, H.; Thaddeus, P.; Turner, B. E.; Yamamoto, S.; Ziurys, L. M. Astrophys. J. 1988, 334, L107. (b) Solomon, P. M.; Jefferts, K. B.; Penzlas, A. A.; Wilson, R. W. Astrophys. J. 1971, 168, L107. (c) Bezard, B.; Marten, A.; Paubert, G. Bull. Am. Astron. Soc. 1993, 25, 1100. (6) Payne, W. A.; Monks, P. S.; Nesbitt, F. L.; Stief, L. J. J. Chem. Phys. 1996, 104, 9808. (7) Holmes, J. L.; Mayer, P. M. J. Phys. Chem. 1995, 99, 1366. (8) Bock, H.; Dammel, R. Chem. Ber. 1987, 120, 1971. Note that the value of 10.58 eV listed in this paper is for the vertical transition, whereas the value given here and in ref 9 is an estimate of the threshold value (i.e., a value that may approximate the adiabatic transition). (9) Lias, S. G.; Liebman, J. F.; Levin, R. D.; Kafafi, S. A. PositiVe Ion Energetics, Version 2.0; NIST Standard Reference Database 19A; NIST: Gaithersburg, MD, 1993. (10) Doughty, A.; Bacskay, G. B.; Mackie, J. C. J. Phys. Chem. 1994, 98, 13546. (11) Horn, M.; Oswald, M.; Oswald, R.; Botschwina, P. Ber. BunsenGes. Phys. Chem 1995, 99, 323. (12) Kuo, S.-C.; Zhang, Z.; Klemm, R. B.; Liebman, J. F.; Stief, L. J.; Nesbitt, F. L. J. Phys. Chem. 1994, 98, 4026. (13) Monks, P. S.; Stief, L. J.; Krauss, M.; Kuo, S. C.; Zhang, Z.; Klemm, R. B. J. Phys. Chem. 1994, 98, 10017. (14) Buckley, T. J.; Johnson, R. D., III.; Huie, R. E.; Zhang, Z.; Kuo, S.-C.; Klemm, R. B. J. Phys. Chem. 1995, 99, 4879. (15) Stief, L. J.; Marston, G.; Nava, D. F.; Payne, W. A.; Nesbitt, F. L. Chem. Phys. Lett. 1988, 147, 570. (16) (a) Nesbitt, F. L.; Monks, P. S.; Scanlon, M.; Stief, L. J. J. Phys. Chem. 1994, 98, 4307 and references therin. (b) Slagle, I. R.; Gutman, D. J. Phys. Chem. 1983, 87, 1818. (17) Stief, L. J.; Nesbitt, F. L.; Payne, W. A.; Kuo, S. C.; Tao, W.; Klemm, R. B. J. Chem. Phys. 1995, 102, 5309. (18) Grover, J. R.; Walters, E. A.; Newman, J. K.; White, M. C. J. Am. Chem. Soc. 1985, 107, 7329 and references therein. (19) Rider, D. M.; Ray, G. W.; Darland, E. J.; Leroi, G. E. J. Chem. Phys. 1981, 74, 1652. (20) Pottie, R. F.; Lossing, F. P. J. Am. Chem. Soc. 1961, 83, 4737. (21) Park, J.-Y.; Gutman, D. J. Phys. Chem. 1983, 87, 1844. (22) Masaki, A.; Tsunashima, S.; Washida, N. J. Phys. Chem. 1995, 99, 13126.